Moon W. Suh

Automatic Recognition of Woven Fabric Patterns by an Artificial Neural Network

A neural network and image processing technology are introduced for classifying woven fabric patterns. An autocorrelation function is used to determine one weave repeat of the fabric. The reflected fabric image is captured and digitized by the computer system. The learning vector quantization algorithm as a learning rule of the artificial neural network enables recognition of woven fabric types more effectively. The results demon strate that three fundamental weave types can be classified accurately, and structural parameters such as yarn spacing, its variance, and the ratio of warp spacing to weft spacing can also be obtained.

A Study of the Shrinkage of Plain Knitted Cotton Fabric, Based on the Structural Changes of the Loop Geometry Due to Yarn Swelling and Deswelling

The structural change of the jersey loop upon yarn swelling is related to the amount of expected laundering shrinkage in cotton jersey fabric by introducing a three-dimen sional loop model. The lengthwise shrinking is explained by the geometry of loop migration and curvature changes upon wetting and drying of the fabric. Also, width wise shrinking is explained by the relationship between wale spacing and yarn diameter. The proposed jersey model explains shrinkage phenomena and also provides a means of estimating stitch length of jersey fabric on the basis of courses per inch, wales per inch, and yarn diameter.The structural change of the jersey loop upon yarn swelling is related to the amount of expected laundering shrinkage in cotton jersey fabric by introducing a three-dimen sional loop model. The lengthwise shrinking is explained by the geometry of loop migration and curvature changes upon wetting and drying of the fabric. Also, width wise shrinking is explained by the relationship between wale spacing and yarn diameter. The proposed jersey model explains shrinkage phenomena and also provides a means of estimating stitch length of jersey fabric on the basis of courses per inch, wales per inch, and yarn diameter.

What is happening to the US textile industry? Reflections on NAFTA and US corporate strategies

The textile and apparel industries in North America have experienced dramatic changes in the past decade. The North American Free Trade Agreement (NAFTA) has prompted the formation of apparel supply networks throughout the Western Hemisphere combining textile industries and retailers in the USA with apparel industries in Mexico to compete against Asian countries. Contrary to the widely acclaimed intent of NAFTA, the increased apparel production in Mexico has not led to a growth for the US textile industry. Instead, the US textile industry has continuously lost ground in global competition, giving up a large portion of its manufacturing. Today, the US textile industry is undergoing negative profits, countless plant closings, layoffs, and eventual bankruptcies. This study analyzes the impact of NAFTA and US textile companies’ corporate strategies on the performance of the textile industry and examines the pending strategic issues for maintaining US textile companies’ competitiveness in global markets.

Sequential Channelling of Mass Variances in Spun Yarn Manufacturing Processes

This research deals with a new method of depicting the irregularity of a spun yarn in terms of the mass density profiles of its roving and sliver. A data acquisition system is introduced for capturing analog signals of fiber strands from an Uster Tester3®. Using the system, a linear relationship can be established between the masses of the fiber assemblies and the amplitudes of the output signals, which are shown to fit a time-series model. In addition, a statistical technique is proposed for separating the input variance and the process variance when a roving is produced from a sliver by the conventional drafting process in ring spinning. A spline method and a cross-spectrum analysis demonstrate that the density profiles of a sliver are channeled down to that of the resulting roving, and suggest a strong possibility for separating and controlling the process variances apart from the input variances.

A New Perspective on Yarn Unevenness: Components and Determinants of General Unevenness

A simple analysis of the local linear density of a yarn yields an equation of its overall variance, which has three components: variance of the number of fibers per cross section, variance of the mean local fiber fineness, and that of the mean parameter of fiber inclination relative to yam axis. Further mathematical analysis of the component variances reveals a set of determining factors: the sequence of the fiber ends along the yam, the distribution of the fiber length, fiber fineness and its irregularity, the irregularity of the fiber configuration relative to the yam axis, and the blend uniformity along the yarn. To help in this analysis, a representation of the yarn, free of any structural hypothesis, is derived from the way the yam emerges from a ring spinning process: a superposition of elementary strips, each resulting from an initial sliver. This represen tation demonstrates that inverse proportionality between the squared CV of the yarn and its mean number of fibers in cross section holds for any yarn, including those idealized by Poissonian or other similar models.

Theoretical and Practical Aspects of Fiber Length Comparisons of Various Cottons

Fiber length is one of the most important physical properties of cotton. Frequently, it is necessary to compare the fiber lengths of various cottons. The commonly used statistical characteristics of a cotton fiber length distribution include the mean fiber length, short fiber content, and upper quartile length. These statistics may be calculated from number-based or weight-based distributions. This paper shows theoretically and experimentally that number-based and weight-based statistics may give opposite rank orders in some cases when they are used to compare cotton fiber lengths.

Estimating Single Cotton Fiber Tensile Properties from the Load-Elongation Curves of Slack Bundles

A method has been developed to estimate single cotton fiber tensile properties from load-elongation curves of slack fiber bundles. The method is applied to bundle load- elongation curves from HVI tests to estimate the averages of fiber breaking strength, elongation, and crimp. The estimated values are compared with single fiber tensile properties obtained from a Mantis® single fiber tester.

Effect of Measurement Principle and Measuring Field on Uniformity Measures of Spun Yarns

The effect of measuring field length on yarn evenness is investigated by comparing the coefficient of variation or CV of the measure obtained from three different sensors with different measurement principles: a capacitance sensor with an 8 mm sensing zone, an optical sensor with a 2 mm sensing zone, and a laser scanner with a 1 mm effective sensing zone. A probabilistic model is developed to predict the different CV values obtained from the different instruments, and the results are compared with experimental values.

TENSILE BEHAVIOR OF SLACK FIBER BUNDLES : THEORY AND APPLICATION TO HVI TESTING

A statistical model for the tensile behavior of a bundle of slack fibers is developed in terms of its constituent single fiber properties. A large amount of data on single fiber tensile properties is obtained by a Mantis® tester. Application of this theory to HVI tensile test results shows much better agreement than other models developed earlier for bundles of straight, equal length fibers.

Static generation and dissipation of polyester continuous filament yarn

The effect of environmental conditions (temperature and relative humidity) and contact conditions (yarn tension and speed) on static generation and dissipation of polyester flat continuous filament yarn, when rubbed against stainless steel was analyzed. A newly developed device, housed in an environmental room, was used to charge the yarn while moving under desired tension. The charge potentials were measured at two different positions in real time. These measurements and previously established exponential relationship permitted the calculation of the initial potential (at the generation point) and a ‘characteristic decay time’, which is a measure of static dissipation. Experimental data showed that temperature, humidity, yarn tension, and yarn speed have significant effects on static generation; while temperature, humidity, and yarn speed yielded statistically significant changes on static dissipation. Anomalous behavior of static charge when measured at a temperature of 35°C, provided a meaningful clue to control the environmental conditions in the textile industry.